DeAI Summit 2026 | Malta 🇲🇹 | 25–26 Nov AI, governance, infrastructure & capital. Who holds the power of AI?
Malta
- Tweets 218
- Following 128
- Followers 43
- Likes 219
Speaker Spotlight: Dr. Gege Gatt
CEO of EBO.AI, lecturer, and specialist examiner at the University of Malta, Dr. Gege Gatt works at the intersection of applied AI, digital ethics, and technology law.
A practical question he brings to DeAI Summit: how do we scale AI while keeping accountability and governance in the loop?
DeAISummit.org
10
A handful of companies are building systems that could shape how billions of people work, communicate, and make decisions.
So where does decentralized AI actually fit?
5 takeaways from our conversation with Mike Robinson 🧵👇
Made with AI
2
1
18
5/ Infrastructure matters more than another hype cycle.
Distributed compute, edge infrastructure, verification, provenance, and better incentive systems are some of the problems Mike believes builders should be thinking about.
Build for the problems that still need solving.
Made with AI
1
3
These questions are exactly why DeAI Summit exists.
Researchers, builders, policymakers, and industry leaders will meet in Malta to debate what decentralized AI should actually look like.
deaisummit.org/
Full conversation with Mike Robinson:
piped.video/Qn3Cui60DjI
Made with AI
2
Who should control the infrastructure behind AI?
Mike Robinson joins The DeAI Podcast to talk about concentrated power, decentralized systems, distributed compute, and what it takes to keep human agency in the picture.
The conversation also gets into the hard part: decentralization still has to solve performance, incentives, accountability, and infrastructure.
Full interview: piped.video/Qn3Cui60DjI
1
2
27
This is the real question behind decentralization.
It’s not just about building smarter AI. It’s about who ends up controlling the infrastructure, access, and power behind it.
Worth thinking about before that control gets too concentrated.
Everyone’s racing to build smarter AI. Who gets to own the infrastructure it runs on?
I spoke with Sabine Roiss about decentralized AI and why we should question where power is accumulating before we’re all renting access from the same few companies.
Full interview in the comments.
1
41
Speaker Spotlight: Dr. @AngeloDalli
Dr. Dalli is the Founder & CEO of UMNAI, where he works on neuro-symbolic AI and explainable, governable decision systems.
His work focuses on a big question for AI: how do we build systems that can reason, explain their decisions, and still remain accountable?
Meet him at DeAI Summit.
DeAISummit.org
1
62
This is the kind of conversation businesses need to be having right now.
Not "AI or no AI," but how to actually use it well without losing judgment, accountability, or people along the way.
Worth watching the full discussion.
My wife’s company told employees to avoid AI.
18 months later, it shut down.
That doesn’t prove causation. But it’s why I take the cost of standing still seriously.
Sabine and I discuss how businesses can put AI to work.
2
46
3 takeaways from our conversation with Sabine Roiss:
1. Decentralized AI is about how power and control are distributed.
2. Governance has to keep up with innovation without leaving safety behind.
3. AI adoption and compute demand are already reshaping businesses and infrastructure.
Full conversation: piped.video/TFOIxOKNK6s
1
1
33
This is exactly why these conversations need to happen now, not later.
AI is moving fast. The rules, safeguards, and accountability around it need to keep up too.
That’s a big part of what we’re building at DeAI Summit
AI doesn’t need bad intentions to create bad outcomes. It only needs access.
Sabine Roiss explains why giving AI control across connected systems makes governance impossible to treat as an afterthought.
Join @DeaiSummit and help shape the rules before the systems shape them.
1
25
Who should control the systems shaping the future of AI?
In Episode 1 of The DeAI Podcast, Sabine Roiss joins us to discuss decentralized AI, governance, AI safety, compute infrastructure, and the risks of concentrating too much power in too few hands.
Watch the full interview:
piped.video/TFOIxOKNK6s
3
21
Reuters reports that Palantir, Nvidia, and Booz Allen are restricting or rethinking how they use frontier AI models over concerns around sensitive data, proprietary information, and retention.
That raises a bigger question:
If you don’t control the model, do you control your data?
As AI moves deeper into enterprise systems, model choice is becoming a question of privacy, trust, and control.
The next AI battleground may not be intelligence.
It may be who keeps the data.
#DeAI #DataOwnership #AISovereignty #EnterpriseAI
2
32
Europe is starting to independently test frontier AI systems for cybersecurity capability and risk.
That raises a bigger governance question:
Who audits the most powerful AI?
As models become more capable, oversight cannot depend only on what developers say about their own systems.
Governance requires the ability to test, verify, and challenge those claims independently.
1
22
Google is investing $15B in AI infrastructure in Finland and has tied part of that expansion to long-term nuclear power.
That points to a bigger shift:
AI is no longer just a compute race.
It is becoming an energy race too.
As models scale, power availability, grid capacity, and long-term energy access are becoming strategic parts of AI infrastructure.
Who powers the AI race?
3
32
Mistral just raised €3B to scale sovereign, open-weight AI and expand its compute infrastructure.
That raises a bigger question: what does AI sovereignty actually require?
A model alone isn't enough. Control over compute, data, infrastructure and deployment matters too.
1
31
Reuters reports that Taiwan is increasingly leveraging its semiconductor leadership as a strategic tool, as the U.S. and Europe push for deeper cooperation and more geographically distributed AI chip manufacturing.
That raises a bigger question:
Who controls the compute?
The AI race isn't only about models. It's about chips, infrastructure, supply chains, and who ultimately has access to the capacity needed to build.
As compute becomes strategic infrastructure, control over it increasingly becomes geopolitical power.
1
3
46